Masters Theses

Rakesh Kumar

Keywords and Phrases

Activity Classification; Chair backrest; Healthcare; Implicit Sensors; Machine Learning; Sedentary Activity

Abstract

"A vast majority of the population spend most of their time in a sedentary position, which potentially makes a chair a huge source of information about a person's daily activity. This information, which often gets ignored, can reveal important health data but the overhead and the time consumption needed to track the daily activity of a person is a major hurdle. Considering this, a simple and cost-efficient sensory system, named Care-Chair, with four square force sensitive resistors on the backrest of a chair has been designed to collect the activity details and breathing rate of the users. The Care-Chair system is considered as an opportunistic environmental sensor that can track each and every activity of its occupant without any human intervention. It is specifically designed and tested for elderly people and people with sedentary job. The system was tested using 5 users data for the sedentary activity classification and it successfully classified 18 activities in laboratory environment with 86% accuracy. In an another experiment of breathing rate detection with 19 users data, the Care-Chair produced precise results with slight variance with ground truth breathing rate. The Care-Chair yields contextually good results when tested in uncontrolled environment with single user data collected during long hours of study."--Abstract, page iii.

Advisor(s)

Das, Sajal K.

Committee Member(s)

Lin, Dan
Jiang, Wei

Department(s)

Computer Science

Degree Name

M.S. in Computer Science

Publisher

Missouri University of Science and Technology

Publication Date

Spring 2016

Pagination

x, 45 pages

Note about bibliography

Includes bibliographical references (pages 42-44).

Rights

© 2016 Rakesh Kumar, All rights reserved.

Document Type

Thesis - Open Access

File Type

text

Language

English

Subject Headings

Chair designSmart materialsHuman-computer interaction

Thesis Number

T 10881

Electronic OCLC #

952595754

Download

Share

 
COinS